In this study, the capability of an electronic nose machine to identify chilled and frozen-thawed chicken meat was investigated. According to the importance of fresh chicken meat consumption in people’s daily diet, recognition of fresh chicken meat from frozen-thawed one is an essential issue. In the mentioned system, eight sensors (MOS type) were applied. The outputs of sensors were preprocessed, then features were extracted, and fuzzy K-nearest neighbors (F-KNN) algorithm was applied for classification. Chicken meats were classified in two classes (fresh and frozen-thawed). Also, each group of fresh and frozen-thawed chicken meat was classified in five classes according to their shelf life (elapsed day after slaughtering/thawing). The obtained results showed that the average amount of accuracy for fresh-chilled chicken meat classification and frozen-thawed one were 95.2 and 94.67%, respectively. Finally, the accuracy of fresh chicken meat with 95.83% demonstrated the high accuracy of it compare with frozen-thawed one. According to the results, high performance of F-KNN algorithm demonstrated that e-nose system can be applied as a rapid, accurate, and nondestructive method for online and automated identification of fresh and frozen-thawed chicken meat.

中文翻译：

---

利用电子鼻机结合模糊KNN识别鲜冷和冻融鸡肉及其保质期

在这项研究中，研究了电子鼻机识别冷藏和解冻鸡肉的能力。根据食用新鲜鸡肉在人们日常饮食中的重要性，从冷冻融化的鸡肉中识别新鲜鸡肉是必不可少的问题。在提到的系统中，应用了八个传感器（MOS型）。对传感器的输出进行预处理，然后提取特征，并应用模糊K最近邻算法（F-KNN）进行分类。鸡肉分为两类（新鲜和冻融）。另外，将新鲜和冷冻解冻的鸡肉的每组根据其保存期限（屠宰/解冻后经过的天数）分为五类。所得结果表明，新鲜冷藏鸡肉分类和冷冻融化鸡肉分类的平均准确度分别为95.2％和94.67％。最后，新鲜鸡肉的准确率达到95.83％，与冷冻融化的鸡肉相比，其准确性更高。根据结果​​，F-KNN算法的高性能表明，电子鼻系统可以作为一种快速，准确且无损的方法，用于在线和自动识别新鲜和冷冻融化的鸡肉。